Against Low-Carb Diets

[teo123]

So, guys, what do you think about my blog-post against low-carb diets? Are the arguments I used there convincing? If not, why?

[brimstoneSalad]

Can you quote them here?

[teo123]

Can you quote them here?

What, you can't open it? Here are the important parts:

Nearly 100 years have passed, and science still hasn't found an answer to the question how can ketogenic diets or fasting possibly help against epilepsy. And more we know about human brain, less plausible it seems that fasting or ketogenic diet help against epilepsy. Anti-diabetic drugs that result in hypoglycemia don't appear to be effective against epilepsy, and neither do the drugs that cause the liver to produce ketone bodies. High-quality studies are, just with any supposed diet cure, difficult or impossible to make because they are very hard to single-blind, yet alone double-blind. Those studies that show it works, of course, nearly always also show it also causes kidney stones (as ketones usually cause), problems with concentration (because human brain is basically made for a high-starch diet and doesn't work well without starch), high cholesterol (because the fats in coconuts and avocados are for the large part saturated fats), and so on.

In the early 1990s, a physician named Robert Coleman Atkins came up with an idea that we should all be eating a diet that's low in carbohydrates, but not necessarily low in protein. That's called Atkins diet. He came up with the idea that starch, along with sugar, is to be blamed for many diseases. In other words, that we should be eating a meat-based diet. Of course, physiologically, blaming starch for diabetes makes about as much sense saying that the cholesterol you eat is to be blamed for heart disease. Most of the cholesterol in your blood is produced by your liver in response to saturated fats, and that's especially true if you have high cholesterol. And some cholesterol is produced in response to fructose, very little of it is the cholesterol you eat. The same is true for glucose: the dangerous spike in glucose levels after eating sugar is not caused by the glucose in sugar, it's caused by the fructose in sugar making your liver behave as if your glucose levels were much lower than they actually are, and mistakenly convert glycogen and protein into glucose. Starch doesn't contain fructose. In fact, of foods that are commonly eaten, only sugar and honey contain large amounts of fructose. Saturated fat also slightly raises the glucose levels in your blood, possibly more than starch.

No human population in the history of civilization has ever been recorded surviving on a vegan diet.

Even if that is true (and not everybody would agree with that), that's using a soft science to contradict a hard one. You don't use some empirical science to contradict mathematics, you don't use astronomy to contradict particle physics, you don't use biology to contradict chemistry, and you don't use anthropology to contradict nutritional science.
Besides, it works both ways. Has there ever been a society which ate a ketogenic diet? You advocate eating coconuts and avocados, does that have a long tradition?

The vegan diet is nutritionally insufficient, lacking not only vitamin B12 but deficient in heme iron and folate (meaning that we should refer to it always as a “vegan diet plus supplements”).

This coming from somebody who advocates a ketogenic diet... Tell me, where would you get vitamin C on a ketogenic diet? From supplements, right?
B12 deficiency has more to do with people living in more sterile environments than with the diet. It's raising around the world, together with the meat consumption. The danger of getting infected with bacteria from plants is way greater than the danger of B12 deficiency, so we kill the bacteria that produce B12 in the plants we eat.
Oh, and, heme iron? As far as I know, according to the mainstream nutrition, heme iron is the bad form of iron that causes colon cancer. Why it is that so many studies show red meat causes cancer, but that white meat and fish doesn't? The main difference between them is, well, the heme iron.
A common claim by those who advocate meat-based diets is that micronutrients are supposedly better absorbed from meat than from plants. Well, I see no reason to think that's the case, other than the bias. It's well-known rabbit meat takes more vitamins to digest than what can be absorbed from it. While it's true that most plants containing calcium contain oxalates (preventing calcium from getting absorbed), calcium in milk gets absorbed... into blood, but not in the bones (for cow's milk contains little to no vitamin K, because cows can synthesize it themselves), and it's possible if not probable that it causes heart disease in humans. Another common argument is that DHA (the form of omega-3-acids found in meat) protects against heart disease better than ALA (the form of omega-3-acids found in plants). But, here is a thing, there is no evidence either protect against heart disease.

A near-vegan diet, in rigorous clinical trials, invariably causes HDL-cholesterol to drop and sometimes raises triglycerides, which are both signs of worsening heart attack risk.

The idea that HDL protects against heart disease is a new one and is not really supported by science. We have a way to, to simplify, turn LDL cholesterol into HDL cholesterol and it's called omega-3-acids. Yet, studies repeatedly fail to prove omega-3-acids protect against heart disease. If omega-3-acids don't help, then moderate drinking (moderate amounts of alcohol raise the HDL but not LDL) doesn't help either, and saturated fats are certainly harmful.
Besides, why do you think clinical trials are the most rigorous form of nutritional science? The most rigorous form of nutritional science is obviously mechanicistic evidence, taking knowledge from harder sciences and applying it to nutritional science. But you clearly don't value that, since mechanicistic evidence is clearly against your claim that starch is to be blamed for various illnesses of today.

Over the last 30 years, as rates of obesity and diabetes have risen sharply in the U.S., the consumption of animal foods has declined steeply: whole milk is down 79%; red meat by 28% and beef by 35%; eggs are down by 13% and animal fats are down by 27%.
Meanwhile, consumption of fruits is up by 35% and vegetables by 20%. All trends therefore point towards Americans shifting from an animal-based diet to a plant-based one, and this data contradict the idea that a continued shift towards plant-based foods will promote health.

And almost the same argument is used by those who claim sugar doesn't cause diabetes, here:

It has become fashionable in recent years to blame sugar for many health problems. However, per capita sugar consumption has actually been falling in the United States since 1999, when bottled water and sugar-free beverages began to edge sodas off the shelf.

So, by the laws of logic, either both of them are right, or none of them are right. I don't know how somebody who has studied nutrition or something related to that can find those arguments convincing, to me it seems worse than the anti-vaccination nonsense.
First of all, how did you acquire those numbers? They seem wildly wrong. I don't know about America, but, in Croatia, in the 1990s (30 years ago), there was a bad food shortage (it was war time). And it was easier to survive on a village because you could grow your own fruits and vegetables. Home-grown fruits and vegetables made most of the diet for most of the people. Eggs being down is believable, since many people in the 1990s were having backyard chickens, but fruits and vegetables being up or red meat being down really isn't plausible.
Second, even if this is true, this ignores the even-more-obvious long-term trend. You know how the emperor Hui Jin was talking If they have no rice, let them eat meat. and was thrown out of power because of insulting his own people like that? Or when Marie Antoinette was talking If they have no bread, let them eat cake.?

Third, even if there were no obvious long-term trend, it would still be the anti-vaccination level nonsense of The rise of autism around the world appears to be correlated with the rise of vaccination.. Correlation does not imply causation. The reason autism is on the rise is that women today give birth when they are older and women who give birth older have a greater chance of having a child with autism or other congenital complications. Similarly, people long ago didn't live long enough to get type-2-diabetes or heart disease, people who would have gotten type-2-diabetes or heart disease simply died of other illnesses while they were young.

There’s the entire Indian subcontinent, where beef is not eaten by the large majority of people, which has seen diabetes explode over the past decade.

And you said you don't like epidemiological studies, but that you only take rigorous studies seriously? How about little consistency here?
Besides, what's your evidence that total saturated fat intake is lower in India than in the US? I see no particular reason to think that's the case. The milk intake is about the same.

[brimstoneSalad]

My first impressions: Contradicting therapeutic consensus on epilepsy isn't a good start. Even if that's true, it's not a good way to open with something so controversial and contrary to medical consensus.

Also, as to its validity, just because we do not know why something works (there's a chance it's due to some allergy or something that an extreme elimination diet just coincidentally avoids) and because it doesn't work in every case (since different cases likely have different causes when the underlying condition itself is poorly understood) doesn't mean it can't be overwhelmingly obvious to therapists that it works in some cases. It's pretty hard to have a placebo effect to make you think your seizures are gone when they aren't.
It doesn't mean something else doesn't work, but it's not hard to determine that it works for some people -- both from the therapists' POV and the patients. Like antidepressants, if you start a therapy and the symptoms disappear again and again and then the symptoms come back when you stop the therapy there's good reason to believe there's some cause there. Could it be placebo? Well for depression placebo is remarkably useful because BELIEF that you'll feel better can affect mood. It's more unlikely for epilepsy.
When one person says he or she did X and some vague and undiagnosed symptom got better, that's a useless anecdote. When thousands of people suffering from something with objective diagnostic criterion start improving after doing X and get worse when they stop doing X, and their doctors can confirm this, that's something different. It's not the gold standard of evidence, like a clinical trial, but it's a good reason to believe something unless or until there's better evidence to the contrary -- and in many cases we're just not going to GET better evidence (e.g. in the case of antidepressants, where the older out-of-patent drugs are likely the most effective and nobody is interested in proving their efficacy anymore: https://slatestarcodex.com/2015/04/30/p ... versities/)

Finally, even if it's true that keto treats epilepsy doesn't mean it's good for other people to do who don't suffer from that. Corpus callosotomy isn't exactly a good idea to do just because it happened to be an effective treatment for epilepsy. Just make that analogy and there's no need to discredit keto as a medical intervention (one with potentially bad side effects!) for people with relatively rare conditions that aren't otherwise well managed by medication.

Anyway, I'll try to read more later, I have to go for now.

[teo123]

contrary to medical consensus

How do you know that's contrary to medical consensus? How do you know what most doctors believe?

it can't be overwhelmingly obvious to therapists that it works in some cases

A few centuries ago, it was obvious to therapists that bloodletting worked.

It's pretty hard to have a placebo effect to make you think your seizures are gone when they aren't.

But they usually aren't gone. They are gone in around 15% or so cases, when they are gone without any medication (and without placebo) in around 7% of cases. Around 50% or so patients report it makes them less common or less severe. That seems like it could easily be due to placebo.

it's a good reason to believe something unless or until there's better evidence to the contrary

Medicine obviously has a horrible track record when you look at its history, so if it claims something that appears not to be based on science, the the right thing to do is to reject it as very likely false.

[Jebus]

How do you know that's contrary to medical consensus? How do you know what most doctors believe?

My guess is he used something called research. You should try it out some time.

A few centuries ago, it was obvious to therapists that bloodletting worked.

There was no such thing as peer reviewed science a few centuries ago. Therapeutic methods were either passed on through tradition or pushed by religious leaders. In the best case, it was based on anecdotes.

Medicine obviously has a horrible track record when you look at its history, so if it claims something that appears not to be based on science, the the right thing to do is to reject it as very likely false.

OK, this was astoundingly stupid, even by your means. Given the fact that you are young, have a fantastic learning tool (Brimstonesalad) at your disposal, and that your IQ is high enough that it has enabled you to master a second language, you should be making some intellectual progress. Your finest moment was a few years ago when you finally conceded that the world is round. Since then you have remained stagnant.

I'm not sure if you are even interested in progressing or if you solely use this forum to promote your own work. Every forum topic somehow seems to end up with you trying to swerve the discussion toward something specific to Croatia that no one else on this forum cares about. You claim to be vegan, but you rarely write anything related to the ethical, environmental, or nutritional benefits of veganism.

We, the moderators, have been extremely patient about your continuous bullshit. You need to start making an effort to improve rather than damage the forum or you will be banned.

[teo123]

My guess is he used something called research

And how do you do proper research on that topic?

In the best case, it was based on anecdotes.

And improperly done experiments (not blinded) aren't much better than anecdotes. Not-blinded experiment is what convinced McArthur Wheeler that lemon juice on face makes him invisible to the cameras.

OK, this was astoundingly stupid, even by your means.

I must admit I don't see why. Medicine does have a terrible track record of convincing itself that treatments which aren't effective or are counter-productive work, even though science can't explain how they would work.

Since then you have remained stagnant.

How do you mean I remained stagnant? I published papers about two sciences, computer science and linguistics.

You claim to be vegan, but you rarely write anything related to the ethical, environmental, or nutritional benefits of veganism.

Well, I am mostly vegan. That doesn't mean that's all I think or talk about.

[brimstoneSalad]

There was no such thing as peer reviewed science a few centuries ago. Therapeutic methods were either passed on through tradition or pushed by religious leaders. In the best case, it was based on anecdotes.

Interestingly, the mechanisms behind fasting and epilepsy were so obvious Hippocrates wrote about it (or one of his school) in ancient Greece.
This same principle is why it's believed ketosis works, it makes your body think it's starving. Usually not a good thing.

Teo, here's a substantial discussion on the topic: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6361831/
Consensus on the point is overwhelming. And there are objective metrics to assess seizures. Perception of frequency of seizures is also not something that's likely to be influenced by placebo, these are very eventful. It's like observing that you used to vomit for a half an hour twice a day, but now you only vomit once a week and it's just a little throw-up in your mouth.

There are some cases where unblinded experiments are very poor in quality, particularly where the effects are subtle or highly subjective, but then there's a level where how obvious the effects are transcends the ability of the average physician to be affected by cognitive bias.
The same in true in fields like physics: when you're measuring something, cognitive bias can easily see the ambiguous measurements rounded up or down, but when you move up a couple orders of magnitude it becomes far too obvious to fuck up by accident.

[teo123]

Interestingly, the mechanisms behind fasting and epilepsy were so obvious Hippocrates wrote about it (or one of his school) in ancient Greece.

It was also obvious to them bloodletting sometimes alleviated the symptoms (hypertension...), so they killed countless patients with that.
A medical treatment dating back to ancient times is not a good argument in its favor. It may even be an argument against it, since there is a bias towards maintaining the tradition.
The fact that some modern papers about the names of places in Croatia cite the Constantine Porphyrogenitus'es suggestions that the name Split comes from Greek for spiny broom ασπαλαθος or, even worse, that Ragusa comes from Latin labina doesn't give credibility to those etymologies, rather it suggests those papers are of low quality.

Consensus on the point is overwhelming.

It doesn't appear to be. ScienceBasedMedicine, one of the most reputable sources about issues of alternative medicine, is against it.

And there are objective metrics to assess seizures.

And ketogenic diet doesn't appear to affect them. This double-blind study found no statistically significant difference in EEG-detectable seizures between children who are fed a keto diet and children who are fed a diet relatively high in glucose, and only marginally statistically significant (p=0.07) difference in subjectively-reported seizures.

Perception of frequency of seizures is also not something that's likely to be influenced by placebo

What do you mean they aren't? This study found that up to 7% of epileptic patients who receive placebo claim to be seizure-free.

[teo123]

Also...

it becomes far too obvious to fuck up by accident.

McArthur Wheeler probably thought that way when deciding not to do a properly controlled and blinded experiment (or even just trying with water instead of just with lemon juice) to determine whether lemon juice makes one invisible to the cameras.